Today's Question:  What does your personal desk look like?        GIVE A SHOUT

Pros and cons of 6 most popular test automation frameworks

  CarolWilliams        2015-11-30 08:14:23       7,401        0    

Not everyone dappling into systems for test automation knows how to leverage the maximum benefit and keep them going in top shape for a long time. That's where test automation frameworks come in – they're basically sets of rules, standards, protocols and guidelines which can be incorporated into practice or simply followed to provide beneficial execution environments for automation test scripts.

 

Why are test automation frameworks important?

 

Test automation frameworks offer many advantages for developers. Providing benefits like ease of scripting, understandability and modularity to process definition, maintenance and re-usability, test automation frameworks are there to make the lives of developers easier.

 

When implementing a test automation framework in your project, take extra care to ensure that every single team member working on different application modules is on the same page – you wouldn't want each of them to implement their own automation models.

 

Here are 6 most popular test automation frameworks together with their pros and cons to help you choose one that perfectly suits your project requirements.

 

1. Module-based testing framework

 

This framework basically divides the entire app into logical and isolated modules. Developers can create a separate and independent test script for each module – put together, those scripts make up for a larger test scrip which represents several modules at once. App modules are generally separated by an abstraction layer, so when developers apply changes in specific application sections, those will have no impact on the module itself.

 

The main advantages of module-based testing framework lie in its main characteristic – a high degree of modularization, which equals lower maintenance cost. Naturally, the framework is highly scalable as well.

 

The most serious problems of the framework are the separations made between test scripts. Developers usually implement test scripts separately for each module and then embed test data onto them. Testing a different set of data will always require some extra work on test scripts – and this can turn out to be quite time-consuming.

 

2. Library architecture testing framework

 

This framework is built on module-based testing framework, but it offers a few more advantages. Working under this framework, developers segregate the application into functions, including those which can be used by its other parts. The regular procedure would be to divide the application under test into test scripts instead.

 

Using the framework, developers can build a library of common functions meant for the application under test. Such libraries can be called within test scripts when needed. Library architectures are the fundamental element of this framework – developers first try to determine common steps and then group them into functions under a library.

 

The main advantage of the framework lies in the fact that it's highly modular, leading developers to easier scalability and cost effective maintenance. When developers create common functions, they do it with the awareness that those can be used by different test scripts over and over again – library architecture framework offers a high degree of re-usability.

 

What are the shortcomings of this framework? Many point to the fact that test data is here lodged into test scripts, so any change in test data will necessarily require changes in test scripts. Moreover, the introduction of libraries makes the framework slightly complicated – perhaps too complex for some developers out there.

 

3. Behavior-driven development framework

 

This framework is an agile testing methodology which helps with automation of functional validations to arrive at a format understandable to everyone, ranging from developers and testers to marketers and business analysts. As such, the framework doesn't assume users to have knowledge of a programming language. Over time, many tools emerged addressing this framework specifically – think Cucumber or Jbehave.

 

The most important advantage of behavior-driven development framework is that it allows a full collaboration between the business and development sides of a project. It forces the business side to show good justification for their prioritization of features. Moreover, the framework addresses the needs and potential behaviors of users at early development stages, examining the system’s behavioral aspects in detail.

 

But this close relationship between technical and non-technical staff can also become a problem – especially if you consider communication issues. Features cannot be built in the framework without first articulating their purpose, the number of people who will use them or what value they will deliver.

 

4. Data-driven testing framework

 

Sometimes developers will need to test a functionality several times with different sets of input data. This is where embedding test data in test scripts becomes extremely unproductive. Developers can however retain test data in a database located outside of test scripts.

 

This framework helps developers to differentiate test script logic and test data, consequently allowing them to store test data in external databases like ODBC repositories, property files, excel files, xml files, text files or CSV files. Data is stored in Key-Value pairs, where keys are used to access and populate data within test scripts.

 

The key advantage of this framework is its impact on script number – developers won't need to write scripts required for every possible test scenario and can arrive at a complete set of scenarios much easier, thus increasing their flexibility.

 

Is there anything negative about the framework? The truth is that it's complex and may require additional effort to devise all test data sources and reading mechanisms. It also requires serious proficiency in the language of test scripts – it's not something for newbies.

 

5. Keyword-driven testing framework

 

This framework is an extension of the data-driven testing framework – it not only segregates test data from scripts, but it also maintains a portion of test script in an external data repository. The latter is known as keywords – they're self-guiding when it comes to actions performed on the application.

 

The advantages of this framework are partly those of the data-driven testing framework, but keyword-driven frameworks won't require developers to have significant scripting knowledge. Generated keywords can be reused many times in a single test case.

 

The framework is not without flaws. Developers should know their way around keyword creation mechanisms – otherwise they won't benefit from this method. Moreover, this kind of framework tends to grow and become increasingly complex as new keywords are introduced – not a great choice for those who value simplicity.

 

6. Hybrid testing framework

 

The very suggestive name of this framework is enough to understand its specific character. Hybrid testing frameworks are basically combinations of at least two different testing frameworks listed above.

 

As you can imagine, the most significant advantage of hybrid testing frameworks is their hybrid nature. They're appreciated by developers because they combine benefits of all associated testing frameworks, helping professionals to leverage all that's best about them in a clear-cut manner.

 

How to choose the right test automation framework?

 

There are several aspects developers look for when choosing the test automation framework for their project. Here are some of the most important ones for you to consider before you make your decision:

 

  • Adaptability – this is a criterion delineating the ease of use, deployment and user learning curve (focusing on the complexity of the framework and development language)

  • Re-usability – a key value of test automation frameworks

  • Reporting – the ability of a framework to produce reports, but also logging, granularity of reporting and ability to transform reports to other storage or format

  • Extensibility – whether the framework can be extended in case it lacks a key feature, how it integrates with other frameworks or components under test, how many libraries are available and whether it's easy to create various scenarios or test cases for scenarios

  • Supported environments and platforms

  • Integration with CI

  • Readability

  • Easy maintenance

  • Support – the community surrounding a particular test automation framework and whether there's existing documentation to help leverage all benefits of the framework.

 

What to remember when designing a test automation framework?

 

It's clear that developing proper frameworks for test automation process is a full time job requiring significant time and effort. When developing a test automation framework, developers should make sure that they first define the goals of test automation. Without having a clear vision of those objectives they cannot define what the test automation is supposed to achieve and consequently won't be able to choose the right framework type.

Before you set out to build a test automation framework for yourself, you should ask the following questions:
What percentage of my application is going to be automated?
What is the shelf of the framework?
Is the application under test stable and if so, to what extent?
Who is the target audience of my testing framework?

Does my target audience possess enough testing skills to use the framework?
What is the projected Return of Investment (ROI) on my automation effort?

Knowing all this in addition to what's best and worst about each of those most popular test automation frameworks will make your choice easier – define your project requirements, check the knowledge of your team and choose the best framework out there to serve your needs.

 

Author's Bio: Carol Williams is part of the team at Florida Oranges. She combines her experience in IT with her love for writing. She is passionate about anything tech and digital-related.

 

Share on Facebook  Share on Twitter  Share on Weibo  Share on Reddit 

  RELATED


No related articles

  0 COMMENT


No comment for this article.